During development of the visual projection of lower vertebrates axons from neural retina form precise, topographically ordered synapses with neurons of the optic tectum. The long-range objective of this research is to understand the mechanisms responsible for this specificity. In light of the inaccessability and complexity of the in vivo situation, the approach outlined here will utilize the observation that retina neurons adhere in vitro to tectal tissue with topographic selectivity that mimics the neuronal projections. As a working hypothesis, it is assumed that elucidation of the biochemical basis for these adhesive preferences will be relevant to synaptic selectivity. The proposed research centers on a plasma membrane protein, ligatin, that was recently discovered in ileal epithelial cells where it serves as a baseplate for external cell-surface proteins. Because ligatin is present on the cell surface as a filament, the proteins bound to it are constrained to a regularly arrayed distribution. Our experiments have shown that ligatin is also a major constituent of plasma membranes in neural retina and that it inhibits the reaggregation of single retinal cells. Here, we test the hypothesis that retinal ligatin attaches the cell-surface molecules that distinguish cells from different regions of the retina and provide for their intercellular adhesive specificities. Possibly differences in the identities and distributions of those molecules along the dorsoventral axes of retina and tectum will be determined, as will the molecules' effects on assays for retinotectal adhesion. In addition, the physicochemical characteristics of retinal ligatin and its distribution on the cell surface will be investigated.